context-search

Search past insights using the local context API and structured filters.

Updated Aug 14, 2025
One-click install
npx skills add https://github.com/kaikojima05/ai-agent-rules --skill context-search
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: context-search
Source: https://github.com/kaikojima05/ai-agent-rules/tree/main/skills/context-search
Command: npx skills add https://github.com/kaikojima05/ai-agent-rules --skill context-search

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When you need an answer, you often repeat research or re-discover past decisions instead of leveraging prior knowledge captured from earlier sessions.

Core Features & Use Cases

  • Knowledge search across insights: Finds relevant past discoveries, solutions, decisions, issues, and caveats using keyword and structured filters.
  • Agent-driven query building: Automatically derives search parameters (q/type/tag/from-to) from the user’s prompt without asking follow-up questions.
  • Readable result synthesis: Organizes results by type, date, and content, including rationales for decisions and follow-ups/relations when present.

Quick Start

Call the skill with the task you are working on so it searches the context dictionary and returns the most relevant past records.

Frequently Asked Questions about context-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search past decisions and rationales to reuse knowledge during coding?

You can retrieve past decisions and rationales by querying a local context API with full-text search, selecting type, tag, and time filters. This surfaces prior solutions and caveats relevant to your current coding workflow without requiring manual research.

How do I automatically build search queries for troubleshooting insights?

Automatically building search queries for troubleshooting insights is done through agent-driven query building, which derives search parameters like keywords, type, tag, and time filters directly from the user's prompt without asking follow-up questions.

What is the best way to surface historical caveats and solutions for a current task?

The best way to surface historical caveats and solutions is to perform a full-text search across a local context dictionary. This retrieves past discoveries and organizes them by type, date, and content to speed up current decisions.

Does the insights API require manual parameter configuration for knowledge search?

No, the insights API does not require manual parameter configuration for knowledge search. The system automatically derives search parameters from the prompt to retrieve relevant past records and presents them in a user-readable structured form.

Can I filter knowledge search results by type and date?

Yes, you can filter knowledge search results by type, tag, and date. The system selects these filters from your prompt to organize past insights, rationales, and follow-ups into a readable structured form.